Method for evaluating the starting timing of geomechanical instability induced by CO2 injection into overpressured reservoirs

Abstract

The method for evaluating the starting time sequence of the geomechanical instability induced by CO2 injection in overpressure reservoirs relates to the technical field of carbon capture, utilization and storage.The present application establishes an initial state of pore pressure and ground stress self-consistency by constructing a two-dimensional fully coupled porous elastoplastic model and using a multi-stage prestress initialization strategy to simulate the natural overpressure background.On this basis, the coupled control effect of injection pressure and well-fault spacing on the time sequence of caprock tensile crack initiation, fault shear activation, plastic accumulation and permeability increase feedback is systematically analyzed.The present application first identifies the "blind fault pressure relief feedback time lag" mechanism in which the caprock tensile crack initiation precedes the effective pressure relief feedback of the fault under the conditions of ultra-low permeability, natural overpressure and short-term strong overload.The present application provides key geomechanical basis for the safe pressure management, well deployment and risk monitoring and early warning in the CO2 geological storage in overpressure reservoirs, and solves the technical problem that it is difficult to determine the competitive instability leading mode in the prior art.

Description

Technical Field

[0001] This invention relates to the field of carbon capture, utilization and storage (CCUS) technology, and in particular to a method for assessing the timing of the onset of geomechanical instability induced by CO2 injection in overpressured reservoirs. Background Technology

[0002] Carbon capture, utilization, and storage (CCUS) and CO2 geological storage are important technological pathways for achieving deep emission reduction and carbon neutrality goals. Deep saline aquifers and depleted oil and gas reservoirs typically have large theoretical storage capacities. However, the safe and efficient injection of CO2 depends not only on the reservoir pore volume and permeability but also on multiple factors such as caprock sealing, fault stability, and the controllability of pore pressure disturbances.

[0003] Geologically, many sedimentary basins can develop natural overpressure during rapid burial, unbalanced compaction, diagenetic fluid release, and hydrocarbon generation. For example, significant overpressure systems have been identified in areas with CO2 geological storage potential, such as the Ordos Basin in China, the Gulf Basin in North America, and the Central Graben in the North Sea, Europe. For reservoirs with high initial pore pressure or anomalous pressure gradients, CO2 injection does not begin with a hydrostatic pressure background but rather involves superimposed engineering pressurization on the existing pore pressure-stress equilibrium. If conventional safe injection experience for hydrostatic reservoirs is still applied, the risk of instability caused by short-term pressure accumulation and rapid decrease in effective stress may be underestimated.

[0004] Existing research on the geomechanical safety of CO2 injection processes has mainly focused on two directions: first, the maximum sustainable injection pressure and fault reactivation; and second, caprock integrity and potential leakage channels. However, for reservoir systems with natural extreme overpressure, ultra-low permeability tight matrix, and blind faults, two key questions remain unclear. First, under short-term, high-overload injection conditions, which will reach the initiation condition first: fault shear activation or caprock fracture initiation? In other words, what is the competitive instability leader mode of the system? Second, if local shear activation of the blind fault is accompanied by increased permeability, can an effective pressure relief feedback be formed in a timely manner, thereby altering the timing of caprock fracture initiation?

[0005] Existing technologies mostly rely on single threshold judgments or independent assessments of fault and caprock risks, lacking synchronous comparison and coupled analysis of the initiation timing of the two instability mechanisms, and failing to reveal the time lag effect of blind fault depressurization feedback. Therefore, developing a method and system capable of comprehensively assessing the initiation timing of geomechanical instability under overpressure reservoir injection is of significant theoretical importance and engineering application value for the safe operation of CCUS projects. Summary of the Invention

[0006] One of the objectives of this invention is to address the shortcomings of the prior art by providing a method for evaluating the timing of the onset of geomechanical instability induced by CO2 injection in overpressured reservoirs.

[0007] This invention addresses the conceptual scenario of "ultra-low permeability dense matrix - natural overpressure - vertical blind fault - short-term strong overload injection". By constructing a high-fidelity numerical model, it analyzes the competitive timing of the caprock and fault responses under pressure disturbance, identifies instability leader modes, and proposes for the first time a "blind fault pressure relief feedback timing lag" mechanism.

[0008] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: a method for evaluating the initiation time of CO2-induced geomechanical instability in overpressured reservoirs, comprising the following steps:

[0009] S1. Construct a two-dimensional fully coupled porous elastoplastic model.

[0010] The model has a horizontal length of 1000 m and a vertical thickness of 200 m; from top to bottom, it consists of a 50 m thick caprock, a 100 m thick target reservoir, and a 50 m thick basement. The fault is a vertical blind fault, penetrating only the reservoir section, without cutting through the caprock or connecting to the open boundary at the top of the model. The injection well is simplified as a point source located in the middle of the reservoir.

[0011] Regarding the solid mechanics boundary conditions, the bottom of the model is set as a fixed constraint, the left and right boundaries are supported by rollers to limit normal displacement, and an equivalent overlying rock load is applied to the top. For the seepage boundary conditions, the left and right boundaries and the bottom are set as non-flow boundaries, and a constant hydrostatic pressure is maintained at the top.

[0012] S2. Establish the self-consistent overpressure initial state of pore pressure-geothermal stress.

[0013] A multi-stage prestress initialization strategy was employed to simulate the natural pressurization process driven by deep hydrocarbon generation:

[0014] Phase I: Establishing the equilibrium state between hydrostatic pressure and initial geostress.

[0015] Phase II: Simulate the hydrocarbon generation and pressurization process by using mass flux at the bottom of the fault, gradually increasing the reservoir pore pressure from the static water background to the target overpressure value (e.g., 42 MPa).

[0016] Phase III: The pore pressure, displacement, and stress field obtained at the end of Phase II are introduced and mechanical rebalancing is performed.

[0017] Phase IV: Based on the self-consistent overpressure initial state described above, pressure-controlled injection and calculations of soil plasticity and permeability evolution are initiated.

[0018] S3. Set the pressure control injection mode.

[0019] To avoid non-physical numerical oscillations caused by the instantaneous change in pressure boundary at the initial injection moment, a one-day linear slope loading is applied to the bottom hole pressure during the injection phase. Let the pre-injection overpressure background pore pressure be p0, and the target injection pressure be P. inj Then the bottom pressure p w (t) is:

[0020] ;

[0021] In the formula, t represents time, specifically the time elapsed since the injection began; d represents the total duration of the ramp loading, which is fixed at 1 day.

[0022] S4. Define the instability initiation criterion.

[0023] 1. Criterion for tensile crack initiation in the caprock: Based on the first principal effective stress σ′1, when σ′1 first exceeds the tensile strength T0 of the caprock (3 MPa in this invention), the condition for tensile crack initiation is considered to have been met.

[0024] .

[0025] 2. Fault Shear Activation Criterion: Based on the Mohr-Coulomb shear yield criterion, when F MC When the value is ≥ 0, the fault zone is considered to have reached the shear yield condition. Yield function F MC Defined as:

[0026] (5);

[0027] Where τ is the shear stress, c is the cohesion, and σ′ n The effective normal stress is given by φ, where φ is the internal friction angle. When F... MC When ≥ 0, the fault zone is considered to have reached the shear yield condition.

[0028] To distinguish the different stages of fault shear activation, the present invention also defines the following events:

[0029] 1. First yielding of the fault: The first time any node in the fault zone touches the Mohr-Coulomb yield surface.

[0030] 2. Accumulation of fault plasticity: The equivalent plastic strain of the fault exceeds the numerical tolerance for the first time (e.g., >1e-6).

[0031] 3. Significant increase in permeability of faults: The permeability of faults shows a significant relative increase for the first time (e.g., k / k0 > 1+ε).

[0032] S5. Set up a penetration rate evolution model.

[0033] To reflect the impact of plastic damage on fluid conductivity, differentiated permeability update functions are used for reservoirs and faults:

[0034] Reservoir: ;

[0035] Fault: ;

[0036] Where, k res k represents the reservoir permeability after the evolution of plastic damage during the injection process. res,0 The reservoir baseline permeability is given by solid.epe, which represents the equivalent plastic strain variable, before injection and before injection. fault k represents the permeability of the fault zone after shear-plastic deformation during the injection process. fault,0 The baseline permeability is the permeability of the pre-fault zone.

[0037] This setting reflects the physical understanding that reservoir brittle damage may lead to a more significant increase in local conductivity, while fault zones, as pre-existing damaged structures, have relatively low permeability sensitivity.

[0038] S6. Perform instability initiation timing analysis.

[0039] Under a unified criterion framework, the initial triggering times of four types of events (caprock fracturing initiation, fault initial yielding, fault plastic accumulation, and significant fault permeability enhancement) are compared. By systematically scanning the injection pressure (e.g., 60-80 MPa) and well-fault spacing (e.g., 100-500 m), an instability initiation response phase diagram is constructed to identify the critical pressure range and instability leader modes.

[0040] S7. Propose safety assessment indicators.

[0041] Based on the analysis results, a "delayed pressure relief feedback mechanism in blind faults" is proposed as a new safety assessment indicator. This mechanism refers to the fact that under conditions of ultra-low permeability, extreme natural overpressure, and short-term strong overload, even if a blind fault undergoes local activation and permeability enhancement, its pressure relief feedback may still lag behind the initiation of the caprock, thus making it difficult to achieve effective pressure redistribution in a timely manner. Therefore, in safety assessments, it is crucial not only to focus on "whether the fault is activated" but also to assess "whether the fault pressure relief is timely enough."

[0042] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0043] 1. First-ever revelation of a competitive instability leader mode: This invention, through the construction of a high-fidelity numerical model and unified event time series comparison, reveals for the first time that in ultra-low permeability, naturally overpressured reservoirs, during high-pressure CO2 injection, caprock fracture initiation can precede fault shear activation and permeability enhancement feedback. This discovery overturns the traditional risk assessment approach that prioritizes fault activation. Under different well spacing conditions, caprock fracture initiation occurs earlier than fault initial yielding, plastic accumulation, and significant permeability enhancement.

[0044] 2. Identification of Short-Term Critical Pressure Range: This invention, through discrete pressure scanning, identifies for the first time a short-term critical pressure range of 68.75-70 MPa for a conceptual model. Compared to an initial overpressure background of 42 MPa, the corresponding critical additional pore pressure is approximately 26.75-28 MPa. This provides a quantitative reference benchmark for pressure management of CO2 injection in overpressured reservoirs. Figure 16 As shown in the instability initiation response phase diagram, the instability initiation of the system is primarily controlled by the injection pressure threshold, rather than by spatial distance.

[0045] 3. A novel mechanism of "time lag in pressure relief feedback from blind faults" is proposed: This invention proposes and verifies for the first time the time lag in pressure relief feedback from blind faults, meaning that even if the fault is activated and permeable, it is insufficient to form an effective pressure relief channel before the caprock fractures. This mechanism provides a theoretical basis for explaining why faults cannot be simply relied upon as a safe pressure relief structure. Even considering the dynamic evolution of reservoir permeability, the trend of caprock fracture initiation remains unchanged.

[0046] 4. Clarifying the limitations of spatial retreat: This invention quantitatively demonstrates that while increasing the well-fault spacing can weaken the fault response, it cannot replace the control of the upper limit of injection pressure, nor can it simultaneously reduce the risk of caprock fracturing. This provides a decision-making basis for prioritizing well location deployment and pressure management in engineering practice.

[0047] 5. Provides a complete model verification system: This invention uses Theis analytical solutions, Terzaghi consolidation theory, and independent trigger benchmark tests to perform layered verification of hydraulic diffusion, hydraulic-mechanical coupling, caprock tensile triggering, and fault shear triggering chains, ensuring the reliability of the model and the robustness of the conclusions. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of a two-dimensional conceptual model, stratigraphic division, location of vertical blind faults, injection point locations, and boundary conditions.

[0049] Figure 2 Schematic diagram for optimized mesh and locally refined areas;

[0050] Figure 3 A schematic diagram of the grid independence verification results: representing the change of absolute pore pressure at the observation point over time;

[0051] Figure 4 This is a schematic diagram showing the change of the maximum first principal effective stress at the bottom of the cover layer over time under a working condition of 80 MPa.

[0052] Figure 5 This is a schematic diagram of Theis's hydrodynamic verification results;

[0053] Figure 6A schematic diagram of the Terzaghi hydraulic-mechanical (HM) benchmark test results: the evolution of surface subsidence over consolidation time;

[0054] Figure 7 A schematic diagram of the Terzaghi HM benchmark test results: the dissipation process of normalized pore pressure at different depths with consolidation time;

[0055] Figure 8 This is a schematic diagram of the benchmark test results for cap layer tension triggering.

[0056] Figure 9 A schematic diagram of the benchmark test results for fault shear triggering and permeability enhancement feedback;

[0057] Figure 10 A schematic diagram comparing the multi-stage initialization and static assignment methods on the vertical displacement profile before injection;

[0058] Figure 11 A comparison of the solution stability of multi-stage initialization and static assignment methods is presented.

[0059] Figure 12 This is a schematic diagram of the spatial distribution of absolute pore pressure along the injection centerline for different well spacings under 80 MPa conditions.

[0060] Figure 13 Permeability evolution contour plots for different well spacing conditions under 80 MPa operating conditions;

[0061] Figure 14 Cloud diagrams of the first principal effective stress of the caprock under different injection pressures and times;

[0062] Figure 15 This is a schematic diagram showing the evolution of the maximum first principal effective stress of the caprock over time under different injection pressures representing well spacing conditions.

[0063] Figure 16 A phase diagram for classifying the instability initiation response in the injection pressure-well spacing parameter space;

[0064] Figure 17 A comparison chart of the first trigger time under different well spacing conditions to unify the judgment criteria;

[0065] Figure 18 A schematic diagram showing the time history of the first principal effective stress of the caprock and the equivalent plastic strain of the fault under the well spacing condition.

[0066] Figure 19 A comparative diagram showing the impact of dynamic evolution of reservoir permeability on the spatial distribution of pore pressure;

[0067] Figure 20A schematic diagram comparing the time history of the maximum first principal effective stress in the caprock under the assumption of dynamic evolution of reservoir permeability and constant permeability.

[0068] Figure 21 This is a schematic diagram comparing the influence of well spacing on fault pressure response with existing research results. Detailed Implementation

[0069] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present invention.

[0070] To balance computational efficiency in engineering-scale simulations with the physical interpretability of complex geomechanical mechanisms, and to fully reflect the coupled control effect of natural overpressure backgrounds and short-term strong overload injection on the initiation timing of caprock and fault instability, this invention innovatively proposes a method and system for evaluating the initiation timing of CO2-induced geomechanical instability in overpressured reservoirs. This method is based on a framework of "two-dimensional fully coupled porous elastoplastic model + multi-stage prestress initialization + unified criterion for caprock fracturing / fault shear." In the model, the hydrodynamic-mechanical bidirectional coupling is characterized by the Biot effective stress principle; fault shear activation and caprock fracturing initiation are characterized by the Mohr-Coulomb criterion and the first principal effective stress criterion, respectively; and fault permeability enhancement feedback is characterized by a permeability-plastic strain evolution model. Combining systematic parameter scanning under the conceptual model with multi-level benchmark testing verification, discrete pressure scanning and a unified timing comparison method are used for instability initiation identification. The results show that, under reasonable parameterization, this method can effectively identify the leading modes of competitive instability, short-term critical pressure ranges, and time lag mechanisms of blind fault depressurization feedback. It provides a comprehensive analytical tool for the safety assessment of CO2 geological storage in overpressured reservoirs, taking into account the competitive instability of caprock-fault formations.

[0071] A method for assessing the onset time of CO2-induced geomechanical instability in overpressured reservoirs includes the following steps:

[0072] S1. Construct a two-dimensional fully coupled porous elastoplastic model, which includes caprock, reservoir, basement and vertical blind fault, and set semi-closed boundary conditions;

[0073] S2. A multi-stage prestress initialization strategy is adopted to simulate the natural pressurization process driven by deep hydrocarbon generation, and to establish an overpressure initial state that is self-consistent with the pore pressure and geostress before injection.

[0074] S3. Set the pressure control injection mode and use linear ramp loading during the injection stage to avoid numerical oscillation;

[0075] S4. Set instability initiation criteria, including the cap layer tensile crack initiation criterion based on the first principal effective stress, and the fault shear activation criterion based on the Mohr-Coulomb yield criterion.

[0076] S5. Under a unified criterion framework, compare the first triggering times of four types of events: caprock rift initiation, fault initial yielding, fault plastic accumulation, and significant fault permeability enhancement, to determine the leading mode of competitive instability.

[0077] S6. Identify the short-term critical pressure range and propose the time lag of blind fault pressure relief feedback as a safety assessment indicator.

[0078] Preferably, in step S1, the governing equations of the two-dimensional fully coupled porous elastoplastic model include the mechanical equilibrium equations under quasi-static conditions and the pore fluid mass conservation equations, and the two-way hydraulic-mechanical coupling is achieved through the Biot effective stress principle equation, wherein:

[0079] The mechanical equilibrium equations under quasi-static conditions are:

[0080] ;

[0081] In the formula, σ is the total stress tensor, ρ is the equivalent density of the porous medium containing fluid, and g is the gravitational acceleration;

[0082] The effective stress principle equation of Biot is:

[0083] ;

[0084] In the formula, σ′ is the effective stress tensor, using the sign convention of positive tension; α B Here, p is the Biot-Willis coefficient, I is the pore pressure, and I is the unit tensor.

[0085] The mass conservation equation for pore fluid is:

[0086] ;

[0087] Among them, S p For water storage, ε v Let be the volumetric strain, k be the permeability tensor, μ be the hydrodynamic viscosity, and Q be the mass source term.

[0088] More preferably, in step S4, the criterion for the initiation of tensile cracks in the caprock is that the first principal effective stress σ′1 exceeds the tensile strength T0 of the caprock for the first time.

[0089] ;

[0090] Where σ′1 is the first principal effective stress, T0 is the tensile strength of the cap layer, and T0 is taken as 3 MPa; when F t When the value is ≥ 0, the cap layer is considered to have reached the condition for tensile crack initiation.

[0091] The fault shear activation criterion is the Mohr-Coulomb yield function:

[0092] (5);

[0093] Where τ is the shear stress, c is the cohesion, and σ′ n The effective normal stress is given by φ, where φ is the internal friction angle; when F MC When ≥ 0, the fault zone is considered to have reached the shear yield condition.

[0094] More preferably, step S4 also includes a permeability evolution model, which describes the changes in conductivity of the reservoir and fault after the accumulation of plastic strain, respectively, and is implemented using the following formula:

[0095] Reservoir: ;

[0096] Fault: ;

[0097] Where, k res k represents the reservoir permeability after the evolution of plastic damage during the injection process. res,0 The reservoir baseline permeability is given by solid.epe, which represents the equivalent plastic strain variable, before injection and before injection. fault k represents the permeability of the fault zone after shear-plastic deformation during the injection process. fault,0 The baseline permeability is the permeability of the pre-fault zone.

[0098] More preferably, in step S2, the multi-stage prestress initialization strategy specifically includes:

[0099] Phase I: Establishing the equilibrium state between hydrostatic pressure and initial geostress;

[0100] Phase II: Simulate hydrocarbon generation and pressurization by using mass flux at the base of the fault to raise the pore pressure to the target overpressure value;

[0101] Phase III: Introduce the pore pressure, displacement, and stress field from the end of Phase II, and perform mechanical rebalancing;

[0102] Phase IV: Initiate the injection phase calculation on the inherited self-consistent overpressure initial state.

[0103] More preferably, in step S6, for the short-term critical pressure range, given an initial overpressure of 42 MPa, it is identified as 68.75-70 MPa, corresponding to a critical additional pore pressure of 26.75-28 MPa (e.g., Figure 16 (As shown).

[0104] More preferably, in step S6, the time lag of the pressure relief feedback of the blind fault means that under conditions of ultra-low permeability, natural extreme overpressure, and short-term strong overload, even if the blind fault undergoes local activation and permeability enhancement, its pressure relief feedback still lags behind the initiation of caprock fractures, making it difficult to form an effective pressure redistribution in a timely manner.

[0105] An evaluation system employing the above method includes:

[0106] The model building module is used to construct a two-dimensional fully coupled porous elastoplastic model;

[0107] The initialization module is used to perform multi-stage prestress initialization and establish a self-consistent overpressure initial state.

[0108] The simulation module is used to set different injection pressures and well-fault spacing to simulate pressure-controlled injection.

[0109] The criteria module includes built-in criteria for caprock tensile fracture initiation and fault shear activation.

[0110] The timing analysis module is used to compare the start times of different instability events and generate an instability initiation response phase diagram (e.g., Figure 16 (as shown)

[0111] The evaluation module is used to assess the safety of storage based on the instability leader mode and critical pressure range.

[0112] Preferably, in the simulation module, the injection well is a point source located in the middle of the reservoir, and the bottom hole pressure p w (t) During the injection phase, a 1-wire linear ramp is used to apply pressure to the target injection pressure P. inj :

[0113] (1);

[0114] Where p0 is the background pore pressure before injection; t represents time, specifically the time elapsed since the start of injection; and d represents the total duration of ramp loading, which is fixed at 1 day.

[0115] The following are specific examples:

[0116] Step 1: Model building and parameter setting.

[0117] First, according to S1 of the present invention, a two-dimensional fully coupled porous elastoplastic model is constructed, the geometry and boundary conditions of which are as follows: Figure 1 As shown. The model is 1000 m long and 200 m high, containing a 50 m caprock, a 100 m reservoir, and a 50 m basement. A vertical blind fault is located in the center, penetrating only the reservoir. The injection well is located in the middle of the reservoir (x=0).

[0118] The main mechanical and hydraulic parameters of each rock stratum and fault zone in the model are shown in Table 1. Key parameters include:

[0119] 1. Cap layer: Young's modulus 20 GPa, cohesion 30 MPa, internal friction angle 35°, tensile strength 3 MPa, initial permeability 1e-26 m².

[0120] 2. Reservoir: Young's modulus 30 GPa, cohesion 15 MPa, internal friction angle 30°, initial permeability 1e-17 m².

[0121] 3. Fault zone: Young's modulus 8 GPa, cohesion 3 MPa, internal friction angle 20°, initial permeability 1e-14 m².

[0122] Table 1

[0123]

[0124] The Biot-Willis coefficient for all rock strata was 0.9, and the hydrodynamic viscosity was 6e-5 Pa·s.

[0125] The model is discretized using a non-uniform triangular mesh, with local refinement applied at the injection point, fault zone, and adjacent areas, such as... Figure 2 As shown. Mesh independence verification ( Figure 3 , Figure 4 This indicates that the optimized mesh scheme can balance computational efficiency while ensuring accuracy.

[0126] Step 2: Establish the initial state of overpressure.

[0127] According to S2 of this invention, a multi-stage prestress initialization strategy is adopted. First, an initial geostress field in equilibrium with hydrostatic pressure and gravity is established. Then, by applying a mass flux at the bottom of the fault, the hydrocarbon generation and pressurization process is simulated, gradually increasing the pore pressure in the central region of the reservoir to an overpressure state of 42 MPa. Subsequently, the pore pressure, stress, and displacement fields under this overpressure state are mechanically rebalanced to eliminate any non-physical initial plastic responses. Finally, a self-consistent overpressure initial state of pore pressure and geostress is obtained, serving as the starting point for subsequent injection simulations.

[0128] like Figure 10 and Figure 11 As shown, multi-stage initialization, compared with the traditional static assignment method, can eliminate the problem of initial displacement jumps and instability in solving, and provides more physically reliable initial conditions for subsequent analysis.

[0129] Step 3: Set the injection plan

[0130] Pressure-controlled injection was employed, with the injection well located at the center of the reservoir. Multiple simulation scenarios were designed, and the system scanned the injection pressure (P).inj The well-to-fault spacing (L) is also considered. This example focuses on the simulation results for injection pressures of 60, 68.75, 70, and 80 MPa, and well-to-fault spacings of 100 m, 150 m, 200 m, and 500 m. A one-day linear slope loading was applied to the bottom hole pressure during the initial injection phase to avoid sudden pressure changes; the calculation formula is as follows:

[0131] (1).

[0132] Step 4: Perform fully coupled simulation and analysis

[0133] Under the initial state established in step 2 and the injection scheme set in step 3, a fully coupled porous elastoplastic simulation is performed to calculate the evolution of pore pressure, stress, displacement, plastic strain and permeability in real time.

[0134] (1) Analysis of pore pressure diffusion and permeability evolution.

[0135] Figure 12 The spatial distribution of pore pressure along the injection centerline at different well spacings (15 days after injection) under an injection pressure of 80 MPa is presented. The results show that when the well-fault spacing L = 100 m (near the fault), the pressure front can be transmitted to the fault location with relatively small attenuation, and the pore pressure increment at the fault is approximately 4.5 MPa. When L = 200 m, the pressure increment transmitted to the fault decreases to approximately 2 MPa. When L = 500 m, the pore pressure at the fault is almost unperturbed. This indicates that increasing the well-fault spacing is an effective spatial retreat strategy to reduce pressure accumulation at the fault.

[0136] Figure 13 The permeability evolution contour plot further supports the above conclusions. In the near-fault condition at L=100 m, local permeability enhancement (k / k0 > 2) occurred in the fault zone and its adjacent shear region due to the accumulation of plastic strain. However, in the condition at L=200 m, the high-permeability area was mainly limited to the vicinity of the injection point, and the permeability of the fault zone remained basically unchanged.

[0137] (2) Stress response of caprock and tensile crack initiation.

[0138] Figure 14 The first principal effective stress cloud diagram of the caprock shows that, under the combined effect of high-pressure injection and the semi-enclosed boundary, a significant tensile stress concentration zone was formed at the bottom of the caprock (especially in the area above the injection point). As the injection time increased, the range of the tensile stress zone expanded.

[0139] Figure 15The evolution of the maximum first principal effective stress at the bottom of the caprock over time under different injection pressures is quantitatively shown. Under the conditions of 70 MPa and 80 MPa, this stress rapidly exceeds the tensile strength threshold of 3 MPa in the initial stage of injection (approximately 0.9 days) (red dashed line in the figure), indicating a risk of caprock tensile crack initiation. However, under the conditions of 60 MPa and 68.75 MPa, the caprock stress consistently fails to reach this threshold.

[0140] (3) Instability initiation timing and competing leader modes.

[0141] To accurately determine the initiation sequence of caprock and fault instability, this invention compares four types of events within a unified criterion framework: caprock fracturing initiation, fault initial yielding, fault plastic accumulation, and significant fault permeability enhancement. The definitions of these criteria are detailed in Table 2.

[0142] Table 2

[0143]

[0144] Figure 17 This is a core result diagram revealing the leading modes of competitive instability. Taking the 80 MPa strong overload condition as an example, this diagram shows the initial triggering times of four types of events under different well-fault spacings (L=100m, 150m, 200m):

[0145] 1. Caprock leader: In all working conditions, the first triggering time of caprock tension initiation (approximately 0.9 days) is earlier than any response on the fault side (yielding, plastic accumulation, permeability enhancement).

[0146] 2. The hysteresis of fault response: The timing of the first yielding of the fault, the accumulation of plasticity and the significant permeability increase are basically coincident or nearly coincident, indicating that the "shear-plasticity-permeability increase" chain is closely coupled, and the conclusion of the caprock leading is not due to the bias of post-treatment standards (such as only looking at permeability increase).

[0147] 3. The effect of spatial setback: As L increases from 100 m to 200 m, the time of the first response of the fault is delayed from about 1.3 days to about 4.3 days (L=150m, yield / increased permeability) or even does not appear within 15 days (L=200m); while the time of crack initiation of the caprock remains at 0.9 days.

[0148] Figure 18 The time history curves further confirm this conclusion, clearly showing the process in which the caprock stress first crosses the threshold, and the fault plastic strain subsequently (or never) initiates.

[0149] Step 5: Identify the critical pressure range and construct the instability phase diagram.

[0150] Based on the calculation results under different injection pressures, this invention plots the instability initiation response phase diagram. Figure 16In the figure, the horizontal axis represents the well-fault distance, and the vertical axis represents the injection pressure. Different colors represent different instability initiation responses within a 15-day injection time window. The results show that:

[0151] 1. Green area (60 MPa, 68.75 MPa): No instability initiation criteria were triggered within 15 days.

[0152] 2. Red area (70-80 MPa): The system enters the instability initiation zone, and in all calculation conditions, it is the precursor to the initiation of caprock tensile cracking.

[0153] 3. No stable zonal structure of "near-field faults being activated first, and far-field caprocks fracturing first" has emerged.

[0154] Therefore, this invention identifies 68.75-70 MPa as the short-term critical pressure range for this conceptual model within a 15-day time window. Relative to the initial overpressure of 42 MPa, the critical additional pore pressure is approximately 26.75-28 MPa.

[0155] Step 6: Propose and verify the "blind fault decompression feedback timing lag" mechanism.

[0156] To test the hypothesis that caprock leaders depend on reservoir permeability enhancement, a comparative simulation was conducted in this embodiment. Figure 19 and Figure 20 The pore pressure distribution and caprock stress evolution were compared under two conditions: considering and not considering the dynamic evolution of reservoir permeability. The results show that even without considering reservoir permeability enhancement, the trend and timing of caprock fracturing do not fundamentally change, and the caprock stress distribution and the time to cross the threshold remain essentially consistent. This demonstrates that the caprock leader is primarily driven by near-wellbore pressure accumulation, rather than entirely dependent on reservoir permeability enhancement.

[0157] Based on the above results, this invention proposes and verifies a "delayed feedback mechanism for pressure relief in blind faults": Under conditions of ultra-low permeability, natural extreme overpressure, and short-term strong overload, the pressure accumulation caused by injection first induces tensile stress concentration and fracturing at the bottom of the caprock. The fault-side response requires the pressure front to propagate to the fault, overcome its frictional strength, and accumulate plastic strain before gradually forming permeability enhancement. Because blind faults do not connect to the top, their local permeability enhancement is insufficient to form systemic pressure relief within the brief window before caprock fracturing. Therefore, even if the fault eventually activates, its pressure relief feedback is "delayed" and insufficient to change the instability leader sequence.

[0158] Step 7: Model Validation.

[0159] To ensure the reliability of the simulation results, the key sub-modules of the model were validated in a layered manner, as shown in Table 3 below.

[0160] Table 3

[0161]

[0162] 1. Hydrodynamic Validation: After degenerating into a pure seepage problem, the radial pressure attenuation calculated by the model is in high agreement with Theis's analytical solution, with a maximum relative error of less than 1.5% (e.g., Figure 5 (As shown).

[0163] 2. Hydraulic-Mechanical Coupling Verification: The model-simulated one-dimensional consolidation settlement process in Terzaghi is consistent with the theoretical solution, with a final relative error of 1.59% and a normalized root mean square error (NRMSE) of 0.69% over the entire time period (e.g., ...). Figure 6 , Figure 7 (As shown).

[0164] 3. Verification of tensile triggering in the caprock: In independent testing, when the first principal effective stress reached the 3 MPa threshold, the equivalent plastic strain immediately began to accumulate, verifying the correct implementation of the tensile cutoff criterion (e.g., Figure 8 (As shown).

[0165] 4. Fault Shear Triggering and Permeability Enhancement Feedback Verification: Independent tests show that at the same moment the Mohr-Coulomb yield function reaches zero, the equivalent plastic strain and the permeability ratio begin to change simultaneously, verifying the temporal consistency of the "yield-plasticity-permeability enhancement" feedback chain (e.g., Figure 9 (As shown).

[0166] The quantitative error indices for model validation are shown in Table 4 below:

[0167] Table 4

[0168]

[0169] The above verifications together constitute the model credibility evidence system of this invention, proving that the model can accurately capture the initiation logic of pressure diffusion, hydraulic-mechanical coupling, and the two types of instability mechanisms.

[0170] Comparative example:

[0171] The present invention also compares the results with existing research. Figure 21 The pressure-distance relationship of this invention was compared with the results obtained by Cao et al. (2025) using the TOUGH-FLAC multiphase flow-geomechanical coupling model. Both results show that well spacing has a significant controlling effect on the pressure response at faults. However, the natural overpressure (initial pore pressure 42 MPa) and short-term strong overload (80 MPa injection pressure) conditions set in this invention lead to a higher order of magnitude of pressure change, further highlighting the necessity of independent assessment of caprock integrity in overpressured reservoirs.

[0172] Furthermore, this invention quantitatively verifies the contribution of key model components to the instability initiation time series analysis through multiple sets of comparative simulations:

[0173] 1. Comparison between multi-stage initialization strategies and traditional static assignment methods (e.g.) Figure 10 , Figure 11 As shown in the figure): Static overpressure assignment can cause significant non-physical displacement jumps before injection, leading to instability in the solution (time step collapse); while the multi-stage prestress initialization strategy proposed in this invention can establish a self-consistent initial state of pore pressure-displacement-stress, effectively eliminating non-physical initial responses and significantly enhancing the numerical stability of fully coupled calculations under high overpressure conditions. This comparison proves the necessity and superiority of the initialization strategy of this invention.

[0174] 2. Comparison between dynamic evolution of reservoir permeability and the assumption of constant permeability (e.g.) Figure 19 , Figure 20 (As shown): To examine whether the caprock leader is overly dependent on reservoir permeability enhancement, this invention compares the spatial distribution of pore pressure and the time history of the maximum first principal effective stress in the caprock under two conditions: considering and not considering the dynamic evolution of reservoir permeability. The results show that even without considering reservoir permeability enhancement, the trend of caprock fracture initiation and the time to cross the threshold do not fundamentally change. This proves that the caprock leader is mainly driven by near-wellbore pressure accumulation, rather than entirely dependent on the reservoir permeability enhancement assumption, thus verifying the robustness of the "blind fault decompression feedback time lag" mechanism.

[0175] The methodology embodied in the above comparative analysis—namely, the step-by-step verification and comparative analysis of complex multi-physics coupling mechanisms—demonstrates the necessity and correctness of gradually introducing multi-physics coupling mechanisms such as hydraulic-mechanical bidirectional coupling, elastoplastic failure criteria, and permeability-plastic strain evolution in this invention.

[0176] In summary, the method and system provided by this invention can effectively identify competitive instability initiation modes, short-term critical pressure ranges, and temporal constraints of blind fault depressurization feedback during high-pressure CO2 injection in overpressured reservoirs. This information is of significant guiding importance for site selection, well location deployment, injection parameter optimization, and risk monitoring and early warning in CO2 geological storage of overpressured reservoirs.

[0177] To facilitate understanding by those skilled in the art of the improvements of this invention over the prior art, some of the accompanying drawings and descriptions have been simplified. The above embodiments are preferred implementations of this invention. In addition, this invention can be implemented in other ways. Any obvious substitutions without departing from the concept of this technical solution are within the protection scope of this invention.

Claims

1. A method for evaluating the timing of CO2-induced geomechanical instability in overpressured reservoirs, characterized in that, Includes the following steps: S1. Construct a two-dimensional fully coupled porous elastoplastic model, which includes caprock, reservoir, basement and vertical blind fault, and set semi-closed boundary conditions; S2. A multi-stage prestress initialization strategy is adopted to simulate the natural pressurization process driven by deep hydrocarbon generation, and to establish an overpressure initial state that is self-consistent with the pore pressure and geostress before injection. S3. Set the pressure control injection mode and use linear ramp loading during the injection stage to avoid numerical oscillation; S4. Set instability initiation criteria, including the cap layer tensile crack initiation criterion based on the first principal effective stress, and the fault shear activation criterion based on the Mohr-Coulomb yield criterion. S5. Under a unified criterion framework, compare the first triggering times of four types of events: caprock rift initiation, fault initial yielding, fault plastic accumulation, and significant fault permeability enhancement, to determine the leading mode of competitive instability. S6. Identify the short-term critical pressure range and propose the time lag of blind fault pressure relief feedback as a safety assessment indicator. In step S1, the governing equations of the two-dimensional fully coupled porous elastoplastic model include the mechanical equilibrium equations under quasi-static conditions and the pore fluid mass conservation equations, and the two-way hydraulic-mechanical coupling is achieved through the Biot effective stress principle equation, wherein: The mechanical equilibrium equations under the quasi-static conditions are as follows: ； In the formula, σ is the total stress tensor, ρ is the equivalent density of the porous medium containing fluid, and g is the gravitational acceleration; The effective stress principle equation of Biot is as follows: ； In the formula, σ′ is the effective stress tensor, using the sign convention of positive tension; α B Here, p is the Biot-Willis coefficient, p is the pore pressure, and I is the unit tensor. The mass conservation equation for the pore fluid is: ； Among them, S p For water storage, ε v Where is volumetric strain, k is the permeability tensor, μ is the hydrodynamic viscosity, Q is the mass source term, and t represents time, specifically the time elapsed since the injection began. In step S4, the criterion for the initiation of caprock tensile cracking is that the first principal effective stress σ′1 first exceeds the tensile strength T0 of the caprock: ； Where σ′1 is the first principal effective stress, T0 is the tensile strength of the cap layer, and T0 is taken as 3 MPa; when F t When the value is ≥ 0, the cap layer is considered to have reached the condition for tensile crack initiation. The fault shear activation criterion is the Mohr-Coulomb yield function: ； Where τ is the shear stress, c is the cohesion, and σ′ n For effective normal stress, φ is the internal friction angle; when F MC When ≥ 0, the fault zone is considered to have reached the shear yield condition.

2. The method for assessing the initiation time of CO2-induced geomechanical instability in overpressured reservoirs according to claim 1, characterized in that: Step S4 also includes a permeability evolution model, which describes the changes in conductivity of the reservoir and fault after the accumulation of plastic strain, and is implemented using the following formula: Reservoir: ; Fault: ; Where, k res k represents the reservoir permeability after the evolution of plastic damage during the injection process. res,0 The reservoir baseline permeability is given by solid.epe, which represents the equivalent plastic strain variable, before injection and before injection. fault k represents the permeability of the fault zone after shear-plastic deformation during the injection process. fault,0 The baseline permeability is the permeability of the pre-fault zone.

3. The method for assessing the initiation time of CO2-induced geomechanical instability in overpressured reservoirs according to claim 1, characterized in that: In step S2, the multi-stage prestress initialization strategy specifically includes: Phase I: Establishing the equilibrium state between hydrostatic pressure and initial geostress; Phase II: Simulate hydrocarbon generation and pressurization by using mass flux at the base of the fault to raise the pore pressure to the target overpressure value; Phase III: Introduce the pore pressure, displacement, and stress field from the end of Phase II, and perform mechanical rebalancing; Phase IV: Initiate the injection phase calculation on the inherited self-consistent overpressure initial state.

4. The method for assessing the initiation time of CO2-induced geomechanical instability in overpressured reservoirs according to claim 1, characterized in that: In step S6, the short-term critical pressure range is identified as 68.75-70 MPa under a given initial overpressure background of 42 MPa, and the corresponding critical additional pore pressure is 26.75-28 MPa.

5. The method for assessing the initiation time of CO2-induced geomechanical instability in overpressured reservoirs according to claim 1, characterized in that: In step S6, the time lag of the pressure relief feedback of the blind fault means that under conditions of ultra-low permeability, natural extreme overpressure, and short-term strong overload, even if the blind fault undergoes local activation and permeability enhancement, its pressure relief feedback still lags behind the initiation of caprock fractures, making it difficult to form an effective pressure redistribution in a timely manner.

6. An evaluation system employing the method described in any one of claims 1-5, characterized in that, include: The model building module is used to construct a two-dimensional fully coupled porous elastoplastic model; The initialization module is used to perform multi-stage prestress initialization and establish a self-consistent overpressure initial state. The simulation module is used to set different injection pressures and well-fault spacing to simulate pressure-controlled injection. The criteria module includes built-in criteria for caprock tensile fracture initiation and fault shear activation. The timing analysis module is used to compare the start times of different instability events and generate an instability initiation response phase diagram. The evaluation module is used to assess the safety of storage based on the instability leader mode and critical pressure range.

7. The evaluation system according to claim 6, characterized in that: In the simulation module, the injection well is a point source located in the middle of the reservoir, and the bottom hole pressure is p. w (t) During the injection phase, a 1-wire linear ramp is used to apply pressure to the target injection pressure P. inj : ； Where p0 is the background pore pressure before injection; d represents the total duration of slope loading, which is fixed at 1 day.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-5.

Images